The increasing prevalence of obesity in the U.S. is responsible for much of the rise in the most common form of diabetes (type 2). This situation demands a careful evaluation of particular weight loss diets with respect to their impact on diabetes risk. We have shown that a popular very low-carbohydrate diet impairs glucose tolerance and promotes insulin resistance in obese rats, despite achievement of weight loss. Our hypothesis is that a very low carbohydrate intake (5% of calories), coupled with the high fat and protein intake that furnishes the other 95% of calories, both decreases the ability of liver and/or muscle to respond to insulin and eventually limits insulin secretion by pancreatic islets, even as the diet promotes weight loss. Protracted insulin resistance and decline in insulin release can lead to eventual diabetes. Since people following very low-carbohydrate diets report high fat intakes (Bravata 2003) our design addresses a real practice among the weight-reducing public. The use of an animal model enables us to obtain the level of carbohydrate restriction and strict control of food intake that has not been realized in human trials addressing the efficacy of this diet (Foster 2003, Shai 2008). We will investigate the effects of a calorie-restricted very low-carbohydrate diet on insulin -regulated gene expression and enzyme activity in liver and muscle, the two major sites of insulin action in order to understand the mechanisms through which the diet produces the observed whole body insulin resistance. In addition, we will study liver and muscle insulin resistance after reintroduction of carbohydrate to assess the persistence of the effects. We will also track the effects of the very low-carbohydrate diet over the course of a year by assessing both muscle and liver insulin resistance and pancreatic insulin secretion; insulin resistance and diminished insulin secretory capacity are the key processes in the development of diabetes. This longitudinal study of insulin resistance and insulin release will allow us to evaluate the diabetogenicity of a diet widely used by individuals already at risk for diabetes.